What Is a System? Parts, Connections, and Purpose

By Pritesh Yadav June 22, 2026 12 min read —

You are surrounded by systems every moment of your life. Your body is a system. The traffic on your street is a system. Your family, your workplace, the economy, the forest near your town, even the thermostat on your wall — all systems. Learning to see them clearly is the single most useful mental skill this book will give you. So let us start at the very beginning and answer the most basic question: what exactly is a system?

The simplest and most respected answer comes from Donella Meadows, an environmental scientist whose 2008 book Thinking in Systems is the foundation of this field. She defined it like this:

Key takeaway: A system is "an interconnected set of elements that is coherently organized in a way that achieves something." Every system — without exception — is built from exactly three things: elements (the parts), interconnections (the relationships and rules), and a function or purpose (the "why").

That short sentence carries a lot of weight. Let's take it apart slowly, because understanding these three pieces — and which of them matters most — is the whole foundation of systems thinking.

The three building blocks of every system

Before we go deep, here are plain definitions of the words we'll use throughout the chapter.

Elements: The visible, countable parts of a system — the players on a football team, the organs in a body, the employees in a company, the trees in a forest.
Interconnections: The relationships, rules, and flows that link the elements and govern how they affect one another — the rules of football, the hormone signals in a body, the policies and routines of a school.
Function (for non-human systems): What a thing-based system does. A thermostat's function is to regulate temperature. It has no intentions; it simply performs a role.
Purpose (for human systems): What a human-run system is organized to achieve. It may be stated openly or left unspoken — and, as we'll see, the real purpose is revealed by behavior, not by mission statements.

Elements are the easy part. They are the things you can point at and count. If you look at a school, the students, teachers, buildings, and textbooks are obvious. Most people, when they look at any situation, see only the elements. That is exactly the trap we want to escape.

Interconnections are harder to see, and they matter more. The relationships hold the elements together and decide how they behave. Here is a striking example from Meadows: take the same 22 people on a pitch, but swap the rules of football for the rules of basketball. Every single element stays the same — yet you now have a completely different system. The parts didn't change; the connections between them did, and that changed everything.

Purpose is the hardest to see, and it matters most of all. Meadows wrote that purpose is "often the most crucial determinant of the system's behavior." Change a system's purpose, and it can transform completely even if every element and every rule stays exactly the same.

Example: Take a football team whose stated purpose is "win the championship." Now change only the purpose to "each player should maximize their individual statistics so scouts notice them." Nothing visible has changed — same players, same coach, same rules — but the team will now behave entirely differently. Players will stop passing, hog the ball, and stop defending. Purpose silently rewires behavior.

A system's true purpose is revealed by behavior, not by words

This is one of the most powerful ideas in the whole book, so let's give it room. Meadows' rule is blunt: "Purposes are deduced from behavior, not from rhetoric or stated goals."

In other words, do not believe what a system says its purpose is. Watch what it actually does over time. Where does the money go? Who gets promoted? What happens when there's pressure and a hard choice must be made? Those decisions reveal the real purpose.

Example: A government proclaims that environmental protection is a top priority — but it assigns almost no budget and no staff to it, and overrules environmental rules whenever they slow down industry. Systemically, environmental protection is not its purpose. Its behavior says otherwise.

Analogy: Judge a system by its habits, not its resolutions. A person who says "I want to get fit" but watches TV every single evening does not, in systems terms, have fitness as their purpose. Their behavior reveals their true purpose: rest and entertainment. Meadows applies this same test to organizations — watch the money flows and the decisions, not the press releases.

Common mistake: Confusing the stated purpose with the actual purpose. Companies announce noble goals — innovation, sustainability, customer obsession — while their incentives and budgets serve something else entirely (often short-term profit). Always ask: "If I only watched its behavior, what would I conclude this system is for?"

A system is more than the sum of its parts

Here is where systems become genuinely magical. When elements connect toward a purpose, the whole can do things that no single part could ever do alone. These surprising new capabilities are called emergent properties.

Emergent property: A capability that arises from the interactions among a system's parts but exists in none of the parts on its own.

One cardiomyocyte — a single heart-muscle cell — cannot pump blood. It can only twitch. But billions of them, connected and coordinated by electrical signals, produce a heartbeat. The cells didn't change. The interaction created something none of them possessed. A heartbeat is emergent. So is consciousness from neurons, wetness from water molecules, and a symphony from individual instruments.

Analogy: A system is like a recipe, not a shopping list. A shopping list — flour, eggs, sugar, butter — is just items sitting next to each other. A recipe adds the interconnections (mix, then heat, in a sequence) and a purpose (make a cake). The cake's taste and texture are emergent: they exist nowhere in any single ingredient. Heat the ingredients separately and you get no cake at all.

Systems versus heaps

The opposite of a system is a heap: a pile of parts with no meaningful connections and no shared purpose. In a heap, the whole equals exactly the sum of its parts — nothing emerges. A pile of sand is a heap. So is a pile of bicycle parts scattered on a garage floor.

The systems thinker Russell Ackoff put it perfectly: "When a system is taken apart, it loses its essential characteristic." A disassembled car is not a slow car or a broken car — it is not a car at all. It's a heap of metal. The "car-ness" lived in the interconnections, and you destroyed them when you took it apart.

Example — Ackoff's "best car" thought experiment (1994): Suppose you gather every car model on the market and hire engineers to find the single best part in each category — the best engine (say, a Rolls-Royce), the best transmission (say, a BMW), the best brakes, the best exhaust (say, a Mercedes) — and then ask mechanics to bolt them all together. You will not get the world's best car. You won't even get a working car, because the parts were never designed to fit one another. The lesson: a system's performance depends on how its parts interact, not on how good each part is in isolation.

The same idea, in reverse: the same parts, arranged differently, become a different system — or no system at all.

The 26 letters of the alphabet can be a meaningless jumble ("hfqwzxmb") or a Shakespearean sonnet. Same elements; the rules of grammar and the purpose of poetry make the difference.
Twenty-two people on a pitch with no rules, roles, or shared goal are a crowd. Add rules, roles, a coach, and a shared purpose, and the same 22 become a team with tactics and coordination.

Analogy: A system versus a heap is like an orchestra versus a room of musicians warming up. In warmup, 80 musicians each play whatever they like — a heap of noise. In performance, the same 80 follow one score and one conductor — a system. The symphony emerges only in the second case, though every musician and instrument is identical in both.

The hidden machinery: stocks, flows, and feedback

To understand why interconnections matter so much, Meadows gives us three more words. We'll explore them deeply in later chapters, but you should meet them now.

Stock: Anything that builds up or drains over time and can be measured at a moment — water in a bathtub, money in a bank, people in a city, trust in a relationship.
Flow: The rate at which a stock fills or empties — water from the tap (inflow), water down the drain (outflow), births, sales, spending.
Feedback loop: A closed chain of cause and effect in which the level of a stock influences the very flows that change it. This is the engine that lets systems self-regulate or self-amplify.

The classic picture is a bathtub. The water already in the tub is the stock. The tap and the drain are the flows.

   tap (inflow)
       |
       v
   +---------------+
   |  ~~~~~~~~~~~   |   <- STOCK (water in tub)
   |  ~~~~~~~~~~~   |
   +-------+-------+
           |
           v
     drain (outflow)

A feedback loop is what happens when the level of the stock reaches back and adjusts a flow. The everyday example is a thermostat.

Example — the thermostat (a balancing feedback loop): Its elements are a temperature sensor, a switch, a furnace or AC unit, and the room air. Its interconnection is a feedback loop: the sensor compares the actual temperature to the set-point; if there's a gap, the switch turns on heating or cooling; the temperature moves back toward the set-point; the gap shrinks. Its purpose is to keep the room at a stable, chosen temperature. The thermostat "senses" the gap between actual and desired, and acts to close it.

  set-point (desired temp)
        |
        v
   compare  --->  gap?  --yes-->  heat/cool
        ^                            |
        |                            v
   measured temp  <----  room temperature changes
        (the loop closes and self-corrects)

This kind of self-correcting, goal-seeking loop is called a balancing loop, and it shows exactly why interconnections — not parts — give a system its behavior. The furnace alone does nothing useful; the loop is what regulates the room.

Why this matters: which lever should you pull?

Now we can answer a question that decides whether you'll be effective or frustrated when you try to change anything. The three components are not equally powerful. Meadows ranks them clearly.

Lever	What you change	How strong is the effect?
Elements	Swap people, replace equipment, add staff	Weakest. Replace every player and you still have a recognizable football team.
Interconnections	Change the rules, incentives, information flows, feedback loops	Stronger. Change the rules from football to basketball and you have a new system.
Purpose	Change what the system is actually for	Strongest. Can transform the system entirely while every element and rule stays the same.

Common mistake: Reaching for the weakest lever first. Most management interventions just swap people or add headcount. But if the structure — the communication, incentives, and feedback loops — is broken, new people will reproduce the same failures. Ackoff's warning applies: improving the parts separately cannot improve the whole when the interactions are poor.

This insight is also why Peter Senge, in his 1990 book The Fifth Discipline, called systems thinking "the fifth discipline" — the one that ties all the others together. His central lesson: most organizational problems (missed deadlines, poor quality, low morale) are symptoms of the system's structure, not failures of individual people. Blaming the person is reaching for the weakest lever. Seeing the structure is where real change begins.

Tip: When something keeps going wrong, resist the urge to ask "Who messed up?" Ask instead: "What structure — what set of rules, incentives, and feedback loops — would reliably produce this result, no matter who is in the seats?" That question moves you from elements to interconnections, where the real leverage lives.

A few traps to avoid

Studying parts in isolation. Dissecting a frog teaches you frog anatomy, but it kills the system — so it can never tell you how a live frog hunts or swims. Breaking a system into parts (called reductionism) is useful for understanding elements but destroys the interconnections that produce the interesting behavior.
Thinking systems must be designed. Forests, weather, and immune responses are systems with no architect. A system needs elements, interconnections, and a function — not a blueprint.
Treating "emergent" as "mysterious." Emergent properties arise from interactions, but they can be studied and predicted once you understand the structure. A heartbeat emerges from heart cells yet is fully explainable by electrophysiology.
Believing more elements is always better. Ackoff's car proves the opposite: more parts, badly connected, make things worse — more features, more people in a meeting, more of anything without coherent interconnections degrades the whole.

Key Takeaways

A system has exactly three components: elements (parts), interconnections (rules and relationships), and a function or purpose (the why) — Donella Meadows.
A system is more than the sum of its parts: it shows emergent properties (a heartbeat, a symphony, a working car) that no single part has alone.
A heap is parts with no connections and no purpose; take a system apart and it collapses into a heap — "a system loses its essential characteristic when taken apart" (Ackoff).
A system's real purpose is revealed by its behavior over time, not by its stated goals — watch the money and the decisions, not the slogans.
The three components are not equal levers: changing elements is weakest, interconnections is stronger, and purpose is strongest.
Most problems are symptoms of structure (interconnections and feedback loops), not of individual people — the foundation of Senge's systems thinking, and of everything that follows in this book.

Continue reading

🔄 Systems Thinking

Systems Thinking vs. Linear Thinking

🔄 Systems Thinking

Stocks and Flows: What Builds Up and What Moves

🔄 Systems Thinking

Why Systems Thinking Matters: Patterns Everywhere